Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Hydranautics Experience
The typical Boron rejection of Hydranautics seawater membranes at nominal test
conditions2 is 92% - 93% depending on element permeability. This corresponds to
about 80% - 86% boron rejection when operated in a typical commercial seawater
system at a flux of 8 gfd (14 lmh) . A comparison of the different Hydranautics SWC
membrane performance is shown in Table 1.
1
Guidelines for drinking-water quality, 3rd ed., Volume1, World Health Organization, 2004
2
Feed TDS = 32,000 mg/L, Feed B = 5 mg/L, Feed Pressure = 800 psi, T = 25C, Rec = 10%, pH = 7
Membrane Flow NaCl Rej B
gpd % Rej %
SWC4+ 6500 99.8 93
SWC3+ 7000 99.83 92
SWC5 9000 99.83 92
Hydranautics routinely performs comparison testing of its elements with those of its
competitors to demonstrate and maintain its technological advantage. A test of six
different types of randomly selected seawater RO elements from the three leading
manufactures confirmed Hydranautics advantage in boron rejection as well as
membrane permeability. Figure 1 below shows the results of this test where each data
point represents the average of all the elements tested for that particular element type.
The data point for the SW30HR-LE corresponds to a single element while the other data
points correspond to an average of multiple elements as follows: TM820-400=6,
SW30HR=4, SWC4+=153, SWC3+=18, SWC5=19.
Seawater Element Comparison
(3.2% NaCl + 5 ppm B, 800 psi, 10% rec, pH = 7)
100
99
98
Boron Rejection (%)
97
96
95
94 SWC5
SWC4+
93
SWC3+
92 SW30HR
91
90 TM820-400 SW30HR LE
89
5,500 6,500 7,500 8,500 9,500
Permeate Flow (gpd)
The lack of charge and small size of the boric acid molecule results in poor rejection by
a reverse osmosis membrane. But at high pH, boron is present as the borate ion,
B(OH)4- ,which is well rejected by the reverse osmosis membrane due to its larger
radius and negative charge. Figure 2 shows estimated nominal boron rejection of the
SWC membrane at 24 C and 30 C as a function of feed pH. Further discussion on how
the RO system designer can take full advantage of this change in rejection as a function
of pH can be found in the next section.
100
99
98
25C 30C
97
Boron Rej (%)
96
95
94
93
92
91
6 7 8 9 10 11
pH
Table 2. Demonstration Sites Utilizing the Boron Rejection Capabilities of Hydranautics SWC
membrane.
The first major seawater commercial system using spiral wound reverse osmosis
membranes to specify a permeate boron limit was the 50,000 m3/day Mediterranean
seawater plant in Larnaca, Cyprus. The plant has been successfully operating since
March of 2001. Since that time, Hydranautics SWC membranes have been producing
potable water with a boron specification of less than 1 parts per million (ppm). The SWC
elements provide optimal performance in Larnacas cutting edge system (designed by
IDE and Hydranautics) which contains a split partial second pass that treats up to 25%
of the overall flow from the first pass when the water temperature ranges between 15-
30 C (59-86 F) to ensure the required quality standards.
Design Considerations
Plants designed with a permeate boron limit usually set that limit at or below the World
Health Organizations 0.5 mg/L. Despite its high boron rejection, a single pass system
equipped with SWC elements alone may not be sufficient to achieve such low levels
when treating typical sea waters, especially at high temperatures. For this reason,
Hydranautics recommends a number of design solutions to further reduce the permeate
boron level. These solutions include:
1. Partial or full second pass RO processing of first pass permeate with pH increase
in the second pass.
2. Using high boron rejecting brackish membranes (ESPAB) in the second pass.
3. Additional permeate polishing with boron selective ion exchange resin
4. First pass pH increase.
5. Elimination of second pass recirculation to the first pass feed.
In a two pass configuration, all or a portion of the permeate from the first pass is pH
adjusted then processed by brackish membranes in a second pass unit (Figure 3). As
with the SWC membrane, the pH increase in the second pass stream results in an
increased ionization rate of boric acid and subsequently leads to higher boron rejection
by the brackish membranes. Due to the risk of scaling at higher pH, the practical pH
limit to the second pass is about 10.5, which corresponds to brackish membranes boron
rejection rate of approximately 95%.
10
9
NaOH 8
3 6
5
1 2
Figure 3. Two pass partial seawater desalination for increased boron rejection.
Depending on boron level requirements and operating conditions, the second pass may
be either a full or partial second pass design. If partial second pass is sufficient to
produce the required boron level in the combined permeate, then application of a split
partial configuration (Figure 4) can improve process economics. When compared to a
conventional two pass design, the split partial configuration can reduce the number of
elements required by 6% to 10%, with a similar reduction in power consumption. As
shown in greater detail in Figure 5, the first pass permeate is collected from both ends
of the first pass pressure vessels. The low salinity (and low boron concentration)
fraction obtained from the feed end is used for blending. The high salinity fraction
obtained from the concentrate end is processed with the second pass unit after pH
adjustment. Due to high feed pH, special attention should be paid to calcium and
magnesium concentrations in the concentrate to avoid scaling.
Figure 4. Two pass split partial seawater desalination for increased boron rejection.
Feed
40,000 ppm TDS
5 ppm Boron First Pass
Brine
30% of
Product
70% of
Product
Second Pass
To further improve the Boron rejection of a two pass RO system, or to remove Boron
from a brackish water source, Hydranautics offers the high boron rejecting ESPA-B.
Built on Hydranautics proven Energy Saving Polyamide (ESPA) technology, the ESPA-
B operates at lower pressures like other low energy brackish elements, but with 33%
lower boron passage. At an elevated pH of 10 and standard brackish test conditions,
the ESPA-B can achieve a boron rejection of 96%.
Ion exchange
The ion exchange system for boron reduction utilizes boron selective resin, which is
regenerated with acid and caustic. In the past, use of ion exchange for boron reduction
was limited mainly to applications in the semiconductor industry. The advantages of ion
exchange resin are low concentration of boron in the effluent (< 0.1 ppm), low power
consumption, and low water losses (high recovery rate). The disadvantages are the
high cost of resin replacement and chemicals for regeneration. Also, because ion
exchange process does not result in any reduction of water salinity, the required level of
permeate salinity must be produced in the first pass unit. Figure 6 shows a typical
integrated system in which ion exchange is used after the first and second pass RO to
further reduce Boron levels.
IX
SWC ESPA
Figure 6. Boron Reduction Using Split Partial Second Pass and Ion Exchange
It has been established that operation at elevated feed pH increases boron rejection.
This approach has been utilized in the second pass RO units where the relatively low
levels of calcium and magnesium allow for operation at higher pH while still avoiding
scaling. But what about an increase in pH to the first pass for the purpose of improved
Boron reduction? Scaling potential in RO seawater systems is not well defined. Early
research postulated operating at a feed pH that would result in negative value of the
Stiff and Davis saturation indicator1, commonly accepted in seawater design for
estimation of calcium carbonate scaling potential. For this reason, commercial
seawater systems operate at reduced feed pH (through acidification) or at native
seawater pH, adding scale inhibitor to prevent potential scaling. Seawater does contain
high concentrations of calcium and magnesium. However, due to its high ionic strength,
the solubility of calcium carbonate and magnesium hydroxide is higher than in low
salinity water. Additionally, in the high ionic strength seawater, ionization of boric acid is
shifted to a lower pH. For this reason, the possibility of increasing boron rejection by
increasing seawater feed pH was explored.
Tests using Pacific seawater spiked with sodium chloride, calcium chloride and
magnesium chloride to simulate concentrate composition were conducted by
Hydranautics to evaluate precipitation potential of calcium carbonate and magnesium
hydroxide in the pH range of 8.1 to 10.42. The laboratory experiments were followed by
test operation of commercial seawater units at high feed pH with good results. The
testing demonstrated the technical and economic feasibility of operating a seawater
system in the feed pH range of 8.3-8.5. Raising the feed pH to such levels could
provide a solution to high boron passage during operating periods of high feed water
temperature, high levels of feed boron, or unexpected deterioration of the membranes
boron rejection.
1
J. W. Strantz, Predicting CaCO3 scaling in seawater RO systems, Technical Proceedings WSIA 10th Annual
Conference and Trade Fair; Water Supply Improvement in the Next Decade, Honolulu, HI, July 25-29, 1982
2
M. Wilf, EVALUATION OF BORON REDUCTION PROCESSES IN RO SEAWATER SYSTEMS, Hydranautics
Internal Technical Report, Oceanside, CA November, 2002.
In a two pass seawater system, it is desirable to recycle the second pass brine stream
back to the first pass feed to reduce overall feed salinity. Unfortunately, in the case of
boron, the opposite is true. Because of the already low feed boron level, recycling the
brine stream may lead to an increase in boron concentration to the RO. For this reason,
elimination of the second pass concentrate recycle, though decreasing the overall
efficiency of the system, can further decrease the final permeate boron level.
1. The ability to design a STANDARD partial or SPLIT partial two pass system.
2. The ability to increase pH to the first AND second pass.
3. A printout which supplies a detailed water analysis (including boron) of the first
and second pass streams as well as front and back streams when designing a
SPLIT partial system.
Disclaimer
This document in no way implies freedom from any patents owned by Hydranautics or others. It is the
responsibility of the designer/customer to be knowledgeable of all relative patents. Hydranautics
assumes no liability or obligation for the information in this document and no guarantees of system or
product performance are expressed or implied by this document unless provided in writing by
Hydranautics.
Lenntech
info@lenntech.com Tel. +31-152-610-900
www.lenntech.com Fax. +31-152-616-289